JP2017507374A - Network service processing method and network service processing apparatus - Google Patents

Abstract

A network service processing method and apparatus are provided. The method divides a field programmable gate array (FPGA) or network processor (NP) into a plurality of mutually isolated service processing resources (110) and includes a first configuration instruction including a first service configuration execution file (120), the service processing resource is configured according to the first service configuration execution file so that the service processing resource services the service packet to be processed (130), and the service transfer is performed with respect to the service packet to be processed. Receiving a forwarding rule to be used when it is to be received (140), receiving a service packet to be processed sent by the user terminal, and processing a service packet to be processed by a service processing resource; Should be processed into service processing resources using forwarding rules Distributing over bis packet (150).

Description

  This application is filed with the Chinese Patent Office on November 26, 2013, which is hereby incorporated by reference in its entirety, Chinese Patent Application No. 20131061381.0 named “NETWORK SERVICE PROCESSING METHOD AND APPARATUS”. And claims the priority of Chinese Patent Application No. 201410185366.0 filed with the Chinese Patent Office on May 4, 2014 and named “NETWORK SERVICE PROCESSING METHOD AND APPARATUS”.

  The present invention relates to the field of communication technology, and in particular, to a network service processing method and a network service processing apparatus.

  Along with the rapid development of server computing power in communication systems, there is an urgent need to migrate increasingly large numbers of network services (such as firewall services) from traditional dedicated device platforms to servers. Currently, as a major trend in the development of next-generation communication networks, network virtualization technology has gradually realized various network service functions using three types of general-purpose devices: servers, switches, and storage devices. It is coming. The network service function is realized by the above-described method, thereby realizing flexible and convenient service deployment and reducing development costs.

  In order to achieve flexible and convenient service deployment, virtualization technology is increasingly used on servers. Currently, a virtual machine manager (hypervisor software) runs on a server to virtualize one physical server to multiple servers, which allows multiple different network service processing softwares that do not affect each other to run on the same server. Will be deployed conveniently.

  However, the solution for realizing network service processing in the prior art has the following disadvantages. That is, if a packet is received after the server has used the hypervisor software, the server's network interface card must send the packet to the virtual switch (vswitch) using the hypervisor software, and the processor in the server Identifies a packet in the vswitch, and the vswitch distributes the packet so that the packet is correctly distributed to different virtual machines (abbreviated as VM), and when the VM receives the packet, the VM performs service processing on the packet. However, if there is a relatively large number of packets, the vswitch can only deliver packets to the correct VM after the processor identifies each received packet, which significantly increases the processor workload and Reduce distribution efficiency when processing packets. In addition, this also indicates that the efficiency with which the processor processes the packet is significantly reduced compared to the efficiency before virtualization.

  Embodiments of the present invention provide a network service processing method and a network service processing apparatus so as to solve the low efficiency problem when a server in the prior art processes a packet.

According to a first aspect, an embodiment of the present invention provides a network service processing method, the method comprising:
Dividing the field programmable gate array FPGA or network processor NP into a plurality of mutually isolated service processing resources;
Receiving a first configuration instruction, wherein the first configuration instruction includes a first service configuration execution file, and receiving the first configuration instruction;
Performing service configuration on the service processing resource according to the first service configuration execution file so that the service processing resource performs service processing on the service packet to be processed;
Receiving a transfer rule used when a service transfer is to be made with respect to a service packet to be processed;
A service to be processed to a service processing resource using a transfer rule so as to receive the service packet to be processed transmitted by the user terminal and perform service processing on the service packet to be processed by the service processing resource Distributing the packets.

In a first possible embodiment, classifying and redirecting a service packet to be processed using forwarding rules and determining a service processing path for the service packet to be processed specifically includes ,
Determining the amount of service processing resources that should be serviced for service packets to be processed using forwarding rules;
Distributing a service packet to be processed to one determined service processing resource when it is determined that the service processing is to be performed on the service packet to be processed; or Determining a processing order for processing service packets to be processed by the service processing resources when it is determined that service processing should be performed on the service packets to be processed by the plurality of service processing resources; Distributing the service packets to be processed to the first service processing resource in the processing order according to the determined processing order.

In connection with the first aspect or the first possible embodiment of the first aspect, in a second possible embodiment, the method comprises:
Collecting statistics on service performance parameters used by the FPGA or NP to perform performance monitoring on the service processing resource when the service processing resource performs service processing on the service packet to be processed;
FPGA or NP sends service performance parameters to a network server, where the service performance parameters send service performance parameters that are used as a basis for the network server to adjust the service load for service processing resources. And further comprising:

In a third possible embodiment, the method is the step of receiving the delivered second configuration instruction when the service processing resource needs to have different service functions, wherein the second configuration instruction is Receiving a second configuration instruction including a second service configuration execution file;
Performing service reconfiguration on the service processing resource according to the second service configuration execution file.

According to a second aspect, one embodiment of the present invention provides a network service processing apparatus, the apparatus comprising a processor and an FPGA or NP,
The processor is configured to divide the FPGA or NP into multiple mutually isolated service processing resources,
The processor receives the first configuration instruction including the first service configuration execution file, and performs service processing on the service packet to be processed by the service processing resource according to the first service configuration execution file. It is further configured to perform service configuration for service processing resources,
The FPGA or NP is configured to receive forwarding rules that are used when a service forwarding should be made with respect to the service packet to be processed,
The FPGA or NP receives the service packet to be processed transmitted by the user terminal and uses the transfer rule to the service processing resource so that the service processing resource performs the service processing on the service packet to be processed. It is further configured to distribute service packets to be processed.

In a first possible embodiment, the processor receives a second configuration instruction delivered by the network server, including a second service configuration execution file, when the service processing resource needs to have different service functions. And
It is further configured to perform service reconfiguration on the service processing resource according to the second service configuration execution file.

In connection with the second possible aspect or the first possible embodiment of the second aspect, in a second possible embodiment, the processor includes a service adaptation module, and the FPGA or NP is a receiver classification module and Service processing resources,
Specifically, the service adaptation module is configured to perform service configuration with respect to the service processing resource according to the first service configuration execution file, and to perform service reconfiguration with respect to the service processing resource according to the second service configuration execution file,
The receiver classification module is specifically configured to set forwarding rules on the receiver classification module,
More specifically, the service processing resource is configured to set a transfer rule on the service processing resource.

In connection with the second possible embodiment of the second aspect, in the third possible embodiment, the receiver classification module specifically services the service packet to be processed according to the forwarding rules. Determine the amount of service processing resources that should be processed,
Distributing service packets to be processed to one determined service processing resource when it is determined that service processing should be performed on service packets to be processed, or multiple When the service processing resource is determined to perform service processing on the service packet to be processed, the processing order for processing the service packet to be processed by the service processing resource is determined and determined. The service packets to be processed are distributed to the first service processing resource in the processing order according to the processed processing order.

In connection with the third possible embodiment of the second aspect, in a fourth possible embodiment, the service processing resource is:
Receive service packets to be processed sent by the receiver classification module;
Perform corresponding service processing on the service packet to be processed,
Determine the next level of service processing resources to be serviced for service packets to be processed according to the forwarding rules;
It is further configured to send service packets to be processed acquired after service processing to the next level of service processing resources.

In connection with the second aspect or the third or fourth possible embodiment of the second aspect, in a fifth possible embodiment, the apparatus further comprises a statistics collection module;
The statistics collection module collects statistics on service performance parameters used to perform performance monitoring on service processing resources when service processing is performed on service packets to be processed by the service processing resources. Is sent to the service adaptation module,
The service adaptation module is further configured to send service performance parameters to the network server, and the service performance parameters are used as a basis for the network server to adjust the service load for the service processing resources.

In a sixth possible embodiment, the device comprises
A transmission module configured to receive a service packet processed by a service processing resource and output the service packet;
A transfer module configured to transfer service packets between the receiver classification module and the service processing resource, and further configured to transfer service packets between the plurality of service processing resources.

  According to the network service processing method and the network service processing apparatus provided in the embodiment of the present invention, a plurality of mutually isolated service processing resources, which are divided FPGAs or NPs, are processed into service packets to be processed. Service configuration is performed according to the configuration instruction for the service processing resource so that service processing is performed on the service processing resource, and the service packet to be processed is used when a service transfer is to be performed on the service packet to be processed A transfer rule is received so that service processing is performed on a service packet that is distributed to the service processing resource according to the transfer rule and the service processing resource is to be processed. Provides fast and flexible distribution of service packets and reduces the efficiency of distribution when the processor processes packets because the processor needs to identify and distribute each received packet when there is a relatively large number of packets The prior art problem of being done is avoided and the efficiency at which the server processes the packets is improved. In addition, the service processing resource can further have different service functions depending on the configuration, so that the function of the service processing resource can be changed at any time, and the service processing resource can be reconfigured. The efficiency with which service processing resources process service packets is improved.

3 is a flowchart of a network service processing method according to an embodiment of the present invention; 1 is a schematic structural diagram of a network service processing device according to an embodiment of the present invention; FIG. 6 is a signaling diagram of a process in which a single service processing resource performs network service processing according to an embodiment of the present invention. FIG. 5 is a signaling diagram of a process in which multiple service processing resources perform network service processing according to an embodiment of the present invention. FIG. 5 is a signaling diagram of a process in which multiple service processing resources perform network service processing according to an embodiment of the present invention. FIG. 6 is another signaling diagram of a process in which multiple service processing resources perform network service processing according to an embodiment of the present invention. FIG. 6 is another signaling diagram of a process in which multiple service processing resources perform network service processing according to an embodiment of the present invention.

  To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be described below with reference to the accompanying drawings in the embodiments of the present invention. Be clear and fully explained. Apparently, the described embodiments are merely a part rather than all of the embodiments of the present invention. All other embodiments obtained by persons of ordinary skill in the art based on the embodiments of the present invention without difficulty will fall within the protection scope of the present invention.

  In order to provide a thorough understanding of the present invention, the following is further described with reference to the accompanying drawings using specific embodiments. These embodiments do not constitute limitations on the embodiments of the present invention.

Embodiment 1
Hereinafter, the network service processing method provided in an embodiment of the present invention will be described in detail with reference to FIG. FIG. 1 is a flowchart of a network service processing method according to this embodiment of the present invention. In this embodiment of the present invention, the following steps are executed by a network service processing apparatus that executes the present network service processing method. As shown in FIG. 1, the present embodiment specifically includes the following steps.

  Step 110: Divide the field programmable gate array FPGA or network processor NP into a plurality of mutually isolated service processing resources.

  Specifically, the network service processing apparatus according to this embodiment of the present invention includes a processor and a field-programmable gate array (abbreviated as FPGA), or the network service processing apparatus includes: It includes a processor and a network processor (abbreviated as NP).

  The processor divides the FPGA or NP into a plurality of mutually isolated service processing resources.

  By way of example and not limitation, the method for isolation is specifically the case where the device includes an FPGA, using PR / PRR splitting technology in one FPGA, and separating the FPGA into multiple isolated wiring sections. Or, if the device includes NP, group different numbers of processor cores into different groups using processor core grouping method in NP to acquire multiple service processing resources by splitting Comprising the steps of:

  Step 120: Receive a first configuration instruction including a first service configuration execution file.

  Specifically, in one example, the network service processing apparatus receives a first configuration command transmitted by a network server, and the first configuration command includes a first service configuration execution file. The first service configuration execution file includes a service configuration file and a service execution file. The service configuration file is specifically a programming file related to the service configuration, and the service execution file is specifically a service process. A programming file related to the performance of a service function by a resource.

  The network service processing apparatus may receive the first configuration command transmitted by another device other than the network server. In the present embodiment of the present invention, an example in which the network service processing apparatus receives the first configuration command transmitted by the network server will be used for description.

  Further, the first configuration instruction further includes service processing resource identification information. The service processing device uses the identification information to determine a service processing resource for which service configuration should be performed. The identification information may be carried in the service configuration file or the identification information is carried separately in the first configuration instruction.

  By way of example and not limitation, the network service processing device performs a service function configuration related to a service processing resource in accordance with the first service configuration execution file, so that the service processing resource performs service processing on a service packet to be processed. As with services such as Network Address Translation (abbreviated NAT) services, Point to Point Protocol over Ethernet (abbreviated PPPoE) services, deep packets, etc. -It has a function to perform inspection (Deep Packet Inspection, DPI) service processing.

  Step 130: Perform a service configuration related to the service processing resource according to the first service configuration execution file so that the service processing is performed on the service packet to be processed.

  Specifically, the network service processing apparatus performs service processing on the service packet to be processed according to the service for which the service processing resource is configured, so that the service configuration related to the service processing resource according to the first service configuration execution file I do.

  It can be appreciated that when service configuration is performed with respect to service processing resources, multiple service processing resources can also be configured for the same service.

  Step 140: Receive a transfer rule used when a service transfer is to be made with respect to a service packet to be processed.

  Specifically, in one example, a network service processing device receives a transfer rule that is configured by a network server and is used when a service transfer is to be performed on a service packet to be processed. In addition, the network server configures transfer rules for the FPGA or NP.

  The network service processing apparatus may receive a transfer rule configured by another device other than the network server. In the present embodiment of the present invention, an example in which a network service processing apparatus receives a transfer rule configured by a network server will be used for description.

  In addition, the forwarding rules include the rule characteristics used when service forwarding should be performed for each service packet for which service processing needs to be performed, and identification of service processing resources that match the rule characteristics accordingly. Information and address information of service processing resources are stored. That is, only service packets that satisfy the rule characteristics can be transferred to the matching service processing resource using the address information of the service processing resource, and the service processing is performed by the service processing resource. Otherwise, discard processing is performed for service packets that do not satisfy the rule characteristics. By way of example and not limitation, a rule is specifically the IP address information, port number information, or VLAN ID identifier of a packet for which service processing needs to be performed, and the IP address can be the source IP address or the destination IP address. It can be an address and is not limited to that.

  In one example, the rule characteristics included in the forwarding rule include a plurality of IP address ranges of service packets for which service processing needs to be executed, identification information of service processing resources that match each IP address range, and service processing resources Address information. It can be seen that forwarding rules can be presented in tabular form. For example, a service packet having a first IP address range matches the identifier of the first service processing resource. In other words, when the IP address of the service packet to be processed received by the network service processing device falls within the first IP address range, the network service processing device is processed according to the address information of the service processing resource. The service packet to be transmitted is transmitted to the first service processing resource that matches the first IP address range.

  In this embodiment of the present invention, the time order relationship may not be limited between step 120, step 130, and step 140.

  Step 150: Receive a service packet to be processed sent by a user terminal, and the service processing resource is processed into a service processing resource using a transfer rule so as to perform service processing on the service packet to be processed Distribute service packets to be processed.

  Specifically, after executing Step 110 to Step 140 described above, the network service processing apparatus receives a service packet to be processed transmitted by the user terminal after performing a plurality of configurations.

  The network service processing device distributes the service packet to be processed using the forwarding rule and sends the service packet to be processed to the corresponding service processing resource, and the service processing resource receives the service packet to be processed Corresponding service processing is performed, thereby improving the efficiency of service packet processing.

Further, the step of distributing the service packet to be processed to the service processing resource using the transfer rule is specifically,
The network service processing device using a transfer rule to determine the amount of service processing resources that should perform service processing on the service packet to be processed.

  The step of determining the amount of service processing resources that should be serviced for the service packet to be processed is specifically a single service packet to be processed to perform the corresponding service processing. This is a step of determining that it is sufficient to have a service processing resource or a plurality of service processing resources are required to jointly perform corresponding service processing.

The step of distributing service packets to be processed to service processing resources is specifically:
When the network service processing device determines that the service processing should be performed on the service packet to be processed by one service processing resource, the network service processing device obtains the address information of the service processing resource from the transfer rule. Obtaining and directly distributing the service packet to be processed to the service processing resource determined according to the address information of the service processing resource, or the network service processing apparatus to the service packet to be processed multiple service processing resources When it is determined that the service processing should be performed, the network service processing apparatus determines the processing order for processing the service packet to be processed by the service processing resource, and transfers the processing according to the determined processing order. From the rules Obtaining address information of a number of service processing resources and distributing service packets to be processed to the first service processing resource according to the address information of the first service processing resource in the processing order.

  According to the network service processing method provided in the embodiment of the present invention, service processing is performed on service packets to be processed by a plurality of mutually isolated service processing resources that are divided FPGAs or NPs. Service processing is performed according to a configuration instruction with respect to a service processing resource, and service processing according to a transfer rule used when a service packet to be processed is to be serviced with respect to a service packet to be processed A forwarding rule is received so that service processing is performed on service packets that are distributed to the resources and the service processing resources are to be processed. Provides fast and flexible distribution of service packets and reduces the efficiency of distribution when the processor processes packets because the processor needs to identify and distribute each received packet when there is a relatively large number of packets The prior art problem of being done is avoided and the efficiency at which the server processes the packets is improved. In addition, the service processing resource can further have different service functions depending on the configuration, so that the function of the service processing resource can be changed at any time, and the service processing resource can be reconfigured. The efficiency with which service processing resources process service packets is improved.

Optionally, in this embodiment of the present invention, the service processing resource includes a step of collecting statistics on the service performance parameter at the same time when performing service processing on the service packet to be processed, Used to perform performance monitoring on service processing resources. By performing performance monitoring on service processing resources, the network operator to which the network server belongs can perform real-time monitoring on the performance of each service processing resource, and at the same time, each service processing to properly allocate network resources. Further adjustment of functions regarding resources can be performed. The specific steps are:
Collecting statistics on service performance parameters used by the FPGA or NP to perform performance monitoring on the service processing resource when the service processing resource performs service processing on the service packet to be processed;
FPGA or NP sends service performance parameters to a network server, where the service performance parameters send service performance parameters that are used as a basis for the network server to adjust the service load for service processing resources. And the steps to do.

  Specifically, when each service processing resource performs corresponding service processing on a service packet to be processed, the FPGA or NP performs performance monitoring on each service processing resource according to the service configuration of each service processing resource. Collect statistics on the service performance parameters used to do it. By way of example and not limitation, service performance parameters specifically include service traffic and service performance status of service processing resources. Specifically, the service performance status refers to the remaining service packet processing capability after each service processing resource processes a service packet.

  In this embodiment of the invention, because the services configured for service processing resources are different, the FPGA or NP collects statistics on service performance parameters corresponding to the service processing resources. For example, statistics on the density of service packets received by each service processing resource, the size of service packets received by each service processing resource, the access density for accessing on-chip / off-chip memory, and the amount of data accessed Are collected according to services configured for service processing resources.

After the service performance parameter for each service processing resource is determined, the FPGA or NP sends the service performance parameter to the network server, which is the basis for the network server to adjust the service load for the service processing resource. Used as.

  Further, the network server determines whether the service processing resource needs to be changed according to the service performance parameter, and when the service processing resource needs to be changed, the network server Performs capacity expansion processing for resources or service reconfiguration processing for service processing resources. In this way, the network operator to which the network server belongs can perform real-time monitoring on the performance of each service processing resource, and at the same time, further function adjustment for each service processing resource can be performed in order to appropriately allocate network resources. It can be carried out.

Optionally, in this embodiment of the present invention, the network server can further perform a step of reconfiguring the service processing resource, so that the function of the service processing resource can be changed at any time, Processing resources can be reconfigured, which improves the efficiency with which each service processing resource processes service packets. The specific steps are:
Receiving a second configuration instruction when the service processing resource needs to have a different service function, the second configuration instruction including a second service configuration execution file, Receiving step;
Performing service reconfiguration on the service processing resource in accordance with the second service configuration execution file.

  Specifically, in one example, the network server can make functional adjustments for each service processing resource according to service performance parameters so as to achieve the objective of appropriately allocating network resources.

  When the service processing resource needs to have a different service function, the network service processing apparatus receives the second configuration instruction distributed by the network server including the second service configuration execution file. The network service processing device reconfigures the service processing resource having the service function according to the second service configuration execution file, thereby changing the function of the service processing resource, and each service processing resource processes the service packet. Efficiency is improved.

  The network service processing apparatus may further receive a second configuration command transmitted by another device other than the network server. In this embodiment of the present invention, an example in which the network service processing apparatus receives the second configuration command transmitted by the network server will be used for description.

  The second service configuration execution file included in the second configuration instruction and the first service configuration execution file included in the first configuration instruction perform different service functions so that the service processing resources have different functions. Can also be used.

  The second service configuration execution file included in the second configuration instruction includes a service configuration file and a service execution file. Specifically, the service configuration file is a programming file related to the service configuration, and the service execution file. Specifically, it is a programming file related to the performance of a service function by a service processing resource.

  The second configuration instruction further includes service processing resource identification information. The service processing device uses the identification information to determine a service processing resource on which service reconfiguration is to be performed. The identification information may be carried in the service configuration file, or the identification information is carried separately in the second configuration command.

  In step 110 of the present embodiment of the present invention, a plurality of mutually isolated service processing resources acquired by division are specifically empty service processing resources that do not have a service processing function or service processing functions. It is an empty service processing resource.

  In one embodiment, when the service processing resource acquired by the division is an empty service processing resource, the service configuration is performed with respect to the empty service processing resource according to steps 120 to 130, and the service with respect to the service processing resource according to this step. Reconfiguration can be performed.

  In another embodiment, when the service processing resource acquired by the division is a non-empty service processing resource, service reconfiguration is performed for the non-empty service processing resource according to step 120 to step 130, and the service processing resource according to this step Service reconfiguration may be performed again.

Embodiment 2.
Accordingly, an embodiment of the present invention further provides a network service processing apparatus. A mounting structure of the present apparatus is shown in FIG. 2, and the present apparatus is configured to realize the network service processing method of the first embodiment of the present invention described above. The apparatus includes a processor 10 and an FPGA 20 or NP20.

  The processor 10 is configured to divide the FPGA or NP into a plurality of mutually isolated service processing resources.

  The processor 10 receives the first configuration instruction including the first service configuration execution file, and performs the service processing on the service packet to be processed by the service processing resource. In accordance with the service processing resource.

  The FPGA 20 or NP20 is configured to receive transfer rules that are used when a service transfer is to be made with respect to the service packet to be processed.

  The FPGA 20 or NP20 receives the service packet to be processed transmitted by the user terminal, and uses the transfer rule to the service processing resource so that the service processing resource performs the service processing on the service packet to be processed. It is further configured to distribute service packets to be processed.

The processor 10 receives the second configuration instruction including the second service configuration execution file when the service processing resource needs to have different service functions,
It is further configured to perform service reconfiguration on the service processing resource according to the second service configuration execution file.

  The processor 10 includes a service adaptation module, and the FPGA 20 or NP20 includes a receiving side classification module and service processing resources.

  Specifically, the service adaptation module 101 is configured to perform service configuration regarding the service processing resource according to the first service configuration execution file, and to perform service reconfiguration regarding the service processing resource according to the second service configuration execution file. .

  Specifically, the receiving side classification module 201 is configured to set a transfer rule on the receiving side classification module.

  Specifically, the service processing resource 202 is configured to set a transfer rule on the service processing resource.

Specifically, the receiving side classification module 201 determines the amount of service processing resources that should perform service processing on the service packet to be processed according to the transfer rule,
Distributing service packets to be processed to one determined service processing resource when it is determined that service processing should be performed on service packets to be processed, or multiple When the service processing resource is determined to perform service processing on the service packet to be processed, the processing order for processing the service packet to be processed by the service processing resource is determined and determined. The service packets to be processed are distributed to the first service processing resource in the processing order according to the processed processing order.

Service processing resource 202 receives the service packet to be processed sent by the receiver classification module,
Perform corresponding service processing on the service packet to be processed,
Determine the next level of service processing resources to be serviced for service packets to be processed according to the forwarding rules;
It is further configured to send service packets to be processed acquired after service processing to the next level of service processing resources.

  The FPGA 20 or NP20 further includes a statistics collection module 203.

  The statistics collection module 203 collects statistics on service performance parameters used to perform performance monitoring on service processing resources when service processing is performed on service packets to be processed by the service processing resources. The parameter is configured to be sent to the service adaptation module.

  The service adaptation module 101 is further configured to send service performance parameters to the network server, and the service performance parameters are used as a basis for the network server to adjust the service load for service processing resources.

The FPGA 20 or NP20 is configured to receive a service packet processed by a service processing resource and output a service packet, and a transmission transmission module 204,
And further includes a transfer module 205 configured to transfer service packets between the receiving side classification module and the service processing resource, and further configured to transfer service packets between the plurality of service processing resources.

  The processor 10 further includes a driver module 102 configured to provide a unified interface for the service adaptation module.

  In this embodiment of the invention, the service adaptation module 101 can be located in a separate entity rather than in a processor.

The FPGA 20 or NP20 manages off-chip memory and provides an off-chip storage controller 206 configured to provide a unified access interface,
A service packet transmission / reception interface 207 configured to receive a service packet transmitted by an external device or to transmit a service packet to the external device;
An instruction transmission interface 208 configured to receive instructions sent by the network server or to receive instructions sent by a service adaptation module located in the driver module or processor.

  Each service processing resource 202 included in the FPGA 20 or NP 20 has a one-to-one correspondence with each service adaptation module 101 included in the processor 10, that is, one service adaptation module 101 is configured and configured with respect to one service processing resource 202. It should be noted that processing is performed. Of course, in actual application, a plurality of service processing resources 202 may correspond to one service adaptation module 101. When the network service processing device includes at least one service processing resource 202, the network server distributes at least one first configuration instruction to at least one service adaptation module 101, each first configuration instruction being a first Includes service configuration executable files. Each service adaptation module 101 performs service configuration for the corresponding service processing resource 202. When performing service configuration, the service adaptation module 202 receives only one service processing resource 202 so that each service processing resource 202 processes only one service for the service packet to be processed when receiving the service packet. Configure one service. In addition, the configuration of the service processing resource 202 can be changed according to a different configuration command transmitted by the network server, for example, a second configuration command according to the above-described embodiment transmitted by the network server. The second configuration instruction includes a second service configuration executable so that the service function of the service processing resource 202 can be changed at any time and the service processing resource 202 can be reconfigured, thereby The processing efficiency of the service processing resource 202 is improved, and it is avoided that the function cannot be changed.

  According to the network service processing apparatus provided in the embodiment of the present invention, service processing is performed on service packets to be processed by a plurality of mutually isolated service processing resources that are divided FPGAs or NPs. Service processing is performed according to a configuration instruction with respect to a service processing resource, and service processing according to a transfer rule used when a service packet to be processed is to be serviced with respect to a service packet to be processed A forwarding rule is received so that service processing is performed on service packets that are distributed to the resources and the service processing resources are to be processed. Provides fast and flexible distribution of service packets and reduces the efficiency of distribution when the processor processes packets because the processor needs to identify and distribute each received packet when there is a relatively large number of packets The prior art problem of being done is avoided and the efficiency at which the server processes the packets is improved. In addition, the service processing resource can further have different service functions depending on the configuration, so that the function of the service processing resource can be changed at any time, and the service processing resource can be reconfigured. The efficiency with which service processing resources process service packets is improved.

Embodiment 3.
To assist in understanding this embodiment of the invention, the following is further described with reference to the accompanying drawings using specific embodiments. This embodiment does not constitute a limitation on the embodiment of the present invention.

  The following will be described in detail with reference to FIG. FIG. 3 is a signaling diagram of a process in which a single service processing resource performs network service processing according to an embodiment of the present invention.

  Specifically, as shown in FIG. 3, an example in which one service processing resource performs service processing on a service packet will be used. In accordance with the description of step 110 to step 140 of the embodiment of the present invention, corresponding configurations are performed with respect to the service processing resource and the receiving side classification module. After the configuration is complete, when the user terminal accesses the network, the receiving classification module receives the first service packet sent by the user terminal, and a unique attribute of the first service packet (e.g., Service packet IP address information, port number information, and VLANID identifier. In this case, the IP address information specifically includes, but is not limited to, source IP address information and destination IP address information). Identify whether it matches the service characteristics of the service (eg, NAT service) in the rule (eg, IP address information, port number information, and VLAN ID identifier of the packet for which the NAT service needs to be executed). If the unique attribute of the first service packet matches the service characteristic of the service in the transfer rule, the receiving classification module should perform service processing on the first service packet according to the transfer rule. (For example, NAT service processing resource) is determined. That is, distribution processing is performed on the first service packet. That is, it is determined that only a single service processing resource is required to perform the service processing corresponding to the first service packet.

  When it is determined that one service processing resource should perform service processing on the first service packet, the address information of the NAT service processing resource is obtained from the forwarding rule. The receiving side classification module performs an encapsulation process on the first service packet in order to acquire the second service packet (specifically, the encapsulation process includes the identification information of the NAT service processing resource and the NAT service). The processing service address information is encapsulated in the header of the first service packet), the second service packet includes NAT service processing resource identification information and NAT service processing resource address information. The receiving side classification module sends the second service packet to the forwarding module, and the forwarding module sends the second service packet to the NAT service processing resource according to the identification information and the address information. The NAT service processing resource receives the second service packet, performs the corresponding NAT service processing on the second service packet, and outputs the processed second service packet.

  At the same time when the NAT service processing resource performs service processing on the second service packet, the statistics collection module collects statistics on the service performance parameter of the NAT service processing resource and sends the service performance parameter to the service adaptation module. The service adaptation module sends service performance parameters to the network server, and the service performance parameters are used as a basis for the network server to adjust the service load for NAT service processing resources. When the NAT service processing resource needs to be changed, the network server performs capacity expansion processing on the NAT service processing resource or performs service reconfiguration processing on the NAT service processing resource according to the service performance parameter.

Embodiment 4.
To assist in understanding this embodiment of the invention, the following is further described with reference to the accompanying drawings using specific embodiments. This embodiment does not constitute a limitation on the embodiment of the present invention.

  The following will be described in detail with reference to FIGS. 4A and 4B. 4A and 4B are signaling diagrams of processes in which a plurality of service processing resources perform network service processing according to an embodiment of the present invention.

  Specifically, as shown in FIGS. 4A and 4B, an example in which two service processing resources perform service processing on a service packet will be used. In accordance with the description of step 110 to step 140 of the embodiment of the present invention, corresponding configurations are performed with respect to the service processing resource and the receiving side classification module. After the configuration is complete, when the user terminal accesses the network, the receiving classification module receives the first service packet sent by the user terminal, and a unique attribute of the first service packet (e.g., Service packet IP address information, port number information, and VLANID identifier. In this case, the IP address information specifically includes, but is not limited to, source IP address information and destination IP address information). Match the service characteristics of multiple services (eg PPPoE and DPI services) in the rule (eg IP address information, port number information, and VLANID identifiers of the packets for which PPPoE and DPI services need to be executed) Identify whether or not. If the unique attribute of the first service packet matches the service characteristics of those multiple services in the forwarding rule, the receiver classification module should perform service processing on the first service packet according to the forwarding rule Service processing resources (eg, PPPoE service processing resources and DPI service processing resources) are determined. That is, distribution processing is performed on the first service packet. That is, it is determined that a plurality of service processing resources are required to perform service processing corresponding to the first service packet.

When it is determined that the two service processing resources should process the two services of the first service packet, the processing order of the PPPoE service processing resource and the DPI service processing resource is further determined, and the PPPoE service processing resource The address information and the address information of the DPI service processing resource are acquired from the transfer rule according to the processing order of the PPPoE service processing resource and the DPI service processing resource. The receiving side classification module performs an encapsulation process on the first service packet in order to obtain the second service packet (specifically, the encapsulation process is performed on the PPPoE to the header of the first service packet). Service processing resource identification information and DPI service processing resource identification information, service processing resource address information, and order information for service processing resources to process services)), the second service packet is , PPPoE service processing resource, service processing resource address information, DPI service processing resource identification information, and order information for the service processing resource to process the service. The receiving side classification module sends the second service packet to the transfer module, and the transfer module sends the second service packet to the first service processing resource indicated by the order information according to the identification information, the address information, and the order information. (E.g., PPPoE service processing resource). The PPPoE service processing resource receives the second service packet and performs a corresponding PPPoE service process on the second service packet. In addition, the PPPoE service processing resource determines that DPI service processing needs to be further performed on the second service packet according to the order information, and then the PPPoE service processing resource acquires the third service packet. In order to do so, the second service packet is encapsulated (specifically, the DPI service processing resource identification information, the DPI service processing resource address information, and the order information are sent to the second service packet). Encapsulating in the header of the packet). The third service packet includes identification information of the DPI service processing resource, address information of the DPI service processing resource, and order information. The PPPoE service processing resource sends the third service packet to the transfer module, and the transfer module sends the third service packet to the DPI service processing resource indicated by the identification information according to the identification information and the address information. The DPI service processing resource receives the third service packet, performs the corresponding DPI service processing on the third service packet, and outputs the processed third service packet.

  At the same time when the PPPoE service processing resource and the DPI service processing resource perform service processing on the second service packet and the third service packet, respectively, the statistics collection module relates to the service performance parameter of the PPPoE / DPI service processing resource. Statistics are collected separately and each service performance parameter is sent separately to the corresponding service adaptation module. Each service adaptation module sends service performance parameters to the network server separately, and the service performance parameters are used as a basis for the network server to adjust the service load for PPPoE / DPI service processing resources. When the PPPoE / DPI service processing resource needs to be changed, the network server performs capacity expansion processing for the PPPoE / DPI service processing resource or reconfigures the service for the PPPoE / DPI service processing resource according to the service performance parameter. Process.

  Since the order information in the second service packet contains the order of just two service processing resources, and one service in the second service packet has already been processed, the DPI will process the processed third service packet. It can be understood that the data is output directly. In actual application, if a service packet needs to be processed by at least two service processing resources, the processing process is similar to that described above, and details are not repeated here.

Embodiment 5.
To assist in understanding this embodiment of the invention, the following is further described with reference to the accompanying drawings using specific embodiments. This embodiment does not constitute a limitation on the embodiment of the present invention.

  The following will be described in detail with reference to FIGS. 5A and 5B. 5A and 5B are another signaling diagrams of processes in which multiple service processing resources perform network service processing according to an embodiment of the present invention.

  In the fourth embodiment, when performing the encapsulation process on the first service packet, the receiving side classification module obtains the second service packet by identifying the identification information of the plurality of service processing resources and the plurality of services. Processing resource order information is encapsulated in the header of the first service packet, so the header of the second service packet is too large, which prevents packet transmission.

  In this embodiment of the invention, the header of the first service packet is simplified to facilitate packet transmission.

  Specifically, as shown in FIGS. 5A and 5B, an example in which two service processing resources perform service processing on a service packet will be used. In accordance with the description of step 110 to step 140 of the embodiment of the present invention, corresponding configurations are performed with respect to the service processing resource and the receiving side classification module. After the configuration is complete, when the user terminal accesses the network, the receiving classification module receives the first service packet sent by the user terminal, and a unique attribute of the first service packet (e.g., Service packet IP address information, port number information, and VLANID identifier. In this case, the IP address information specifically includes, but is not limited to, source IP address information and destination IP address information). Match the service characteristics of multiple services (eg PPPoE and DPI services) in the rule (eg IP address information, port number information, and VLANID identifiers of the packets for which PPPoE and DPI services need to be executed) Identify whether or not. If the unique attribute of the first service packet matches the service characteristics of those multiple services in the forwarding rule, the receiver classification module should perform service processing on the first service packet according to the forwarding rule Service processing resources (eg, PPPoE service processing resources and DPI service processing resources) are determined. That is, distribution processing is performed on the first service packet. That is, it is determined that a plurality of service processing resources are required to perform service processing corresponding to the first service packet.

When it is determined that the two service processing resources should process the two services of the first service packet, the processing order of the PPPoE service processing resource and the DPI service processing resource is further determined, and the PPPoE service processing resource The address information and the address information of the DPI service processing resource are acquired from the transfer rule according to the processing order of the PPPoE service processing resource and the DPI service processing resource. The receiving side classification module performs the encapsulation process on the first service packet to acquire the second service packet (specifically, the encapsulation process is performed according to the service processing order of the two service processing resources). The identification information and address information of the first service processing resource in the service processing order, for example, the identification information and address information of the PPPoE service processing resource are encapsulated in the header of one service packet). The second service packet includes PPPoE service processing resource identification information and address information. The receiving side classification module sends the second service packet to the transfer module, and the transfer module transfers the second service packet to the PPPoE service processing resource. The PPPoE service processing resource performs the corresponding PPPoE service processing for the second service packet, and the unique attribute of the second service packet matches the service characteristics of multiple services (for example, DPI service) in the service forwarding rule table Identify whether to do. If the unique attribute of the second service packet matches the service characteristics of those multiple services in the service forwarding rule table, the PPPoE service processing resource further needs to perform service processing on the second service packet. A service processing resource (for example, DPI service processing resource) is determined, and the DPI acquires identification information and address information of the DPI service processing resource from the transfer rule. If the unique attribute of the second service packet does not match the service characteristics of those services in the service forwarding rule table, the second service packet is output. The PPPoE service processing resource performs an encapsulation process on the second service packet in order to acquire the third service packet (specifically, the encapsulation process includes identification information and address information of the DPI service processing resource). First
The third service packet includes identification information and address information of the DPI service processing resource. The PPPoE service processing resource sends the third service packet to the transfer module, and the transfer module sends the third service packet to the DPI service processing resource indicated by the identification information according to the identification information and the address information. The DPI service processing resource receives the third service packet, the DPI service processing resource performs the corresponding DPI service processing on the third service packet, and the unique attribute of the third service packet is in the service forwarding rule table. Identify if it matches the service characteristics of multiple services. If the unique attribute of the third service packet matches the service characteristics of those multiple services in the service forwarding rule table, the DPI service processing resource further needs to perform service processing on the third service packet. The service processing resource to be determined is determined. If the unique attribute of the third service packet does not match the service characteristics of the plurality of services in the service forwarding rule table, the third service packet is output.

  At the same time when the PPPoE service processing resource and the DPI service processing resource perform service processing on the second service packet and the third service packet, respectively, the statistics collection module relates to the service performance parameter of the PPPoE / DPI service processing resource. Statistics are collected separately and each service performance parameter is sent separately to the corresponding service adaptation module. Each service adaptation module sends service performance parameters to the network server, and the service performance parameters are used as a basis for the network server to adjust the service load for PPPoE / DPI service processing resources. When the PPPoE / DPI service processing resource needs to be changed, the network server performs capacity expansion processing for the PPPoE / DPI service processing resource or reconfigures the service for the PPPoE / DPI service processing resource according to the service performance parameter. Process.

  Since the transfer rules are configured separately for each service processing resource, each service processing resource determines the service processing resource that will process the service packet next, thereby reducing the number of bytes in the header of the service packet, It can be understood that the flexible transmission of the service is facilitated and the processing efficiency of each service processing resource is improved.

  In actual application, if a service packet needs to be processed by at least two service processing resources, the processing process is similar to that described above, and details are not repeated here.

  One skilled in the art can further appreciate that the units and algorithm steps may be implemented by electronic hardware, computer software, or a combination thereof, in combination with the examples described in the embodiments disclosed herein. . In order to clearly describe the compatibility between hardware and software, the above generally describes the configuration and steps of each example according to function. Whether each function performs hardware or software depends on the particular application and design constraints of the technical solution. Those skilled in the art can use various methods to implement the described functions for each application, but such embodiments should not be considered beyond the scope of the present invention.

  The method or algorithm steps described in the embodiments disclosed herein may be implemented by hardware, software modules executed by a processor, or a combination thereof. Software modules can be random access memory (RAM), memory, read only memory (ROM), electrically programmable ROM, electrically erasable ROM, registers, hard disk, removable disk, CD-ROM, or in the art It can be placed on any other known form of storage medium.

  In the foregoing specific embodiments, the objects, technical solutions, and beneficial effects of the present invention are described in detail. It should be understood that the foregoing descriptions are merely specific embodiments of the present invention and are not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, or improvement made without departing from the spirit and principle of the present invention shall fall within the protection scope of the present invention.

10 processor
20 FPGA or NP
101 Service adaptation module
102 Driver module
201 Receiver classification module
202 Service processing resources
203 Statistics collection module
204 Transmission module
205 Transfer module
206 Off-chip storage controller
207 Service packet transmission / reception interface
208 Command transmission interface

  This application is filed with the Chinese Patent Office on November 26, 2013, which is hereby incorporated by reference in its entirety, Chinese Patent Application No. 20131061381.0 named “NETWORK SERVICE PROCESSING METHOD AND APPARATUS”. And claims the priority of Chinese Patent Application No. 201410185366.0 filed with the Chinese Patent Office on May 4, 2014 and named “NETWORK SERVICE PROCESSING METHOD AND APPARATUS”.

  The present invention relates to the field of communication technology, and in particular, to a network service processing method and a network service processing apparatus.

  Along with the rapid development of server computing power in communication systems, there is an urgent need to migrate increasingly large numbers of network services (such as firewall services) from traditional dedicated device platforms to servers. Currently, as a major trend in the development of next-generation communication networks, network virtualization technology has gradually realized various network service functions using three types of general-purpose devices: servers, switches, and storage devices. It is coming. The network service function is realized by the above-described method, thereby realizing flexible and convenient service deployment and reducing development costs.

  In order to achieve flexible and convenient service deployment, virtualization technology is increasingly used on servers. Currently, a virtual machine manager (hypervisor software) runs on a server to virtualize one physical server to multiple servers, which allows multiple different network service processing softwares that do not affect each other to run on the same server. Will be deployed conveniently.

  However, the solution for realizing network service processing in the prior art has the following disadvantages. That is, if a packet is received after the server has used the hypervisor software, the server's network interface card must send the packet to the virtual switch (vswitch) using the hypervisor software, and the processor in the server Identifies a packet in the vswitch, and the vswitch distributes the packet so that the packet is correctly distributed to different virtual machines (abbreviated as VM), and when the VM receives the packet, the VM performs service processing on the packet. However, if there is a relatively large number of packets, the vswitch can only deliver packets to the correct VM after the processor identifies each received packet, which significantly increases the processor workload and Reduce distribution efficiency when processing packets. In addition, this also indicates that the efficiency with which the processor processes the packet is significantly reduced compared to the efficiency before virtualization.

  Embodiments of the present invention provide a network service processing method and a network service processing apparatus so as to solve the low efficiency problem when a server in the prior art processes a packet.

According to a first aspect, an embodiment of the present invention provides a network service processing method, the method comprising:
Dividing the field programmable gate array FPGA or network processor NP into a plurality of mutually isolated service processing resources;
Receiving a first configuration instruction, wherein the first configuration instruction includes a first service configuration execution file, and receiving the first configuration instruction;
Performing service configuration on the service processing resource according to the first service configuration execution file so that the service processing resource performs service processing on the service packet to be processed;
Receiving a transfer rule used when a service transfer is to be made with respect to a service packet to be processed;
A service to be processed to a service processing resource using a transfer rule so as to receive the service packet to be processed transmitted by the user terminal and perform service processing on the service packet to be processed by the service processing resource Distributing the packets.

In a first possible embodiment, classifying and redirecting a service packet to be processed using forwarding rules and determining a service processing path for the service packet to be processed specifically includes ,
Determining the amount of service processing resources that should be serviced for service packets to be processed using forwarding rules;
Distributing a service packet to be processed to one determined service processing resource when it is determined that the service processing is to be performed on the service packet to be processed; or Determining a processing order for processing service packets to be processed by the service processing resources when it is determined that service processing should be performed on the service packets to be processed by the plurality of service processing resources; Distributing the service packets to be processed to the first service processing resource in the processing order according to the determined processing order.

In connection with the first aspect or the first possible embodiment of the first aspect, in a second possible embodiment, the method comprises:
Collecting statistics on service performance parameters used by the FPGA or NP to perform performance monitoring on the service processing resource when the service processing resource performs service processing on the service packet to be processed;
FPGA or NP sends service performance parameters to a network server, where the service performance parameters send service performance parameters that are used as a basis for the network server to adjust the service load for service processing resources. And further comprising:

In a third possible embodiment, the method is the step of receiving the delivered second configuration instruction when the service processing resource needs to have different service functions, wherein the second configuration instruction is Receiving a second configuration instruction including a second service configuration execution file;
Performing service reconfiguration on the service processing resource according to the second service configuration execution file.

According to a second aspect, one embodiment of the present invention provides a network service processing apparatus, the apparatus comprising a processor and an FPGA or NP,
The processor is configured to divide the FPGA or NP into multiple mutually isolated service processing resources,
The processor receives the first configuration instruction including the first service configuration execution file, and performs service processing on the service packet to be processed by the service processing resource according to the first service configuration execution file. It is further configured to perform service configuration for service processing resources,
The FPGA or NP is configured to receive forwarding rules that are used when a service forwarding should be made with respect to the service packet to be processed,
The FPGA or NP receives the service packet to be processed transmitted by the user terminal and uses the transfer rule to the service processing resource so that the service processing resource performs the service processing on the service packet to be processed. It is further configured to distribute service packets to be processed.

In a first possible embodiment, the processor receives a second configuration instruction delivered by the network server, including a second service configuration execution file, when the service processing resource needs to have different service functions. And
It is further configured to perform service reconfiguration on the service processing resource according to the second service configuration execution file.

In connection with the second possible aspect or the first possible embodiment of the second aspect, in a second possible embodiment, the processor includes a service adaptation module, and the FPGA or NP is a receiver classification module and Service processing resources,
Specifically, the service adaptation module is configured to perform service configuration with respect to the service processing resource according to the first service configuration execution file, and to perform service reconfiguration with respect to the service processing resource according to the second service configuration execution file,
The receiver classification module is specifically configured to set forwarding rules on the receiver classification module,
More specifically, the service processing resource is configured to set a transfer rule on the service processing resource.

In connection with the second possible embodiment of the second aspect, in the third possible embodiment, the receiver classification module specifically services the service packet to be processed according to the forwarding rules. Determine the amount of service processing resources that should be processed,
Distributing service packets to be processed to one determined service processing resource when it is determined that service processing should be performed on service packets to be processed, or multiple When the service processing resource is determined to perform service processing on the service packet to be processed, the processing order for processing the service packet to be processed by the service processing resource is determined and determined. The service packets to be processed are distributed to the first service processing resource in the processing order according to the processed processing order.

In connection with the third possible embodiment of the second aspect, in a fourth possible embodiment, the service processing resource is:
Receive service packets to be processed sent by the receiver classification module;
Perform corresponding service processing on the service packet to be processed,
Determine the next level of service processing resources to be serviced for service packets to be processed according to the forwarding rules;
It is further configured to send service packets to be processed acquired after service processing to the next level of service processing resources.

In connection with the second aspect or the third or fourth possible embodiment of the second aspect, in a fifth possible embodiment, the apparatus further comprises a statistics collection module;
The statistics collection module collects statistics on service performance parameters used to perform performance monitoring on service processing resources when service processing is performed on service packets to be processed by the service processing resources. Is sent to the service adaptation module,
The service adaptation module is further configured to send service performance parameters to the network server, and the service performance parameters are used as a basis for the network server to adjust the service load for the service processing resources.

In a sixth possible embodiment, the device comprises
A transmission module configured to receive a service packet processed by a service processing resource and output the service packet;
A transfer module configured to transfer service packets between the receiver classification module and the service processing resource, and further configured to transfer service packets between the plurality of service processing resources.

  According to the network service processing method and the network service processing apparatus provided in the embodiment of the present invention, a plurality of mutually isolated service processing resources, which are divided FPGAs or NPs, are processed into service packets to be processed. Service configuration is performed according to the configuration instruction for the service processing resource so that service processing is performed on the service processing resource, and the service packet to be processed is used when a service transfer is to be performed on the service packet to be processed A transfer rule is received so that service processing is performed on a service packet that is distributed to the service processing resource according to the transfer rule and the service processing resource is to be processed. Provides fast and flexible distribution of service packets and reduces the efficiency of distribution when the processor processes packets because the processor needs to identify and distribute each received packet when there is a relatively large number of packets The prior art problem of being done is avoided and the efficiency at which the server processes the packets is improved. In addition, the service processing resource can further have different service functions depending on the configuration, so that the function of the service processing resource can be changed at any time, and the service processing resource can be reconfigured. The efficiency with which service processing resources process service packets is improved.

3 is a flowchart of a network service processing method according to an embodiment of the present invention; 1 is a schematic structural diagram of a network service processing device according to an embodiment of the present invention; FIG. 6 is a signaling diagram of a process in which a single service processing resource performs network service processing according to an embodiment of the present invention. FIG. 5 is a signaling diagram of a process in which multiple service processing resources perform network service processing according to an embodiment of the present invention. FIG. 5 is a signaling diagram of a process in which multiple service processing resources perform network service processing according to an embodiment of the present invention. FIG. 6 is another signaling diagram of a process in which multiple service processing resources perform network service processing according to an embodiment of the present invention. FIG. 6 is another signaling diagram of a process in which multiple service processing resources perform network service processing according to an embodiment of the present invention.

The purpose of embodiments of the present invention, technical solutions, and to the advantages more clearly below, the technical solutions in the embodiments of the connection with the present invention in the accompanying drawings in the embodiments of the present invention theory to Akira. Apparently, the described embodiments are merely a part rather than all of the embodiments of the present invention. All other embodiments obtained by persons of ordinary skill in the art based on the embodiments of the present invention without difficulty will fall within the protection scope of the present invention.

In order to provide a thorough understanding of the present invention, the following is further described with reference to the accompanying drawings using specific embodiments. These embodiments are not intended to constitute a limitation against this onset bright.

Embodiment 1
Hereinafter, the network service processing method provided in an embodiment of the present invention will be described in detail with reference to FIG. FIG. 1 is a flowchart of a network service processing method according to this embodiment of the present invention. In this embodiment of the present invention, the following steps are executed by a network service processing apparatus that executes the present network service processing method. As shown in FIG. 1, the present embodiment specifically includes the following steps.

  Step 110: Divide the field programmable gate array FPGA or network processor NP into a plurality of mutually isolated service processing resources.

  Specifically, the network service processing apparatus according to this embodiment of the present invention includes a processor and a field-programmable gate array (abbreviated as FPGA), or the network service processing apparatus includes: It includes a processor and a network processor (abbreviated as NP).

  The processor divides the FPGA or NP into a plurality of mutually isolated service processing resources.

  By way of example and not limitation, the method for isolation is specifically the case where the device includes an FPGA, using PR / PRR splitting technology in one FPGA, and separating the FPGA into multiple isolated wiring sections. Or, if the device includes NP, group different numbers of processor cores into different groups using processor core grouping method in NP to acquire multiple service processing resources by splitting Comprising the steps of:

  Step 120: Receive a first configuration instruction including a first service configuration execution file.

  Specifically, in one example, the network service processing apparatus receives a first configuration command transmitted by a network server, and the first configuration command includes a first service configuration execution file. The first service configuration execution file includes a service configuration file and a service execution file. The service configuration file is specifically a programming file related to the service configuration, and the service execution file is specifically a service process. A programming file related to the performance of a service function by a resource.

  The network service processing apparatus may receive the first configuration command transmitted by another device other than the network server. In the present embodiment of the present invention, an example in which the network service processing apparatus receives the first configuration command transmitted by the network server will be used for description.

  Further, the first configuration instruction further includes service processing resource identification information. The service processing device uses the identification information to determine a service processing resource for which service configuration should be performed. The identification information may be carried in the service configuration file or the identification information is carried separately in the first configuration instruction.

  By way of example and not limitation, the network service processing device performs a service function configuration related to a service processing resource in accordance with the first service configuration execution file, so that the service processing resource performs service processing on a service packet to be processed. As with services such as Network Address Translation (abbreviated NAT) services, Point to Point Protocol over Ethernet (abbreviated PPPoE) services, deep packets, etc. -It has a function to perform inspection (Deep Packet Inspection, DPI) service processing.

  Step 130: Perform a service configuration related to the service processing resource according to the first service configuration execution file so that the service processing is performed on the service packet to be processed.

  Specifically, the network service processing apparatus performs service processing on the service packet to be processed according to the service for which the service processing resource is configured, so that the service configuration related to the service processing resource according to the first service configuration execution file I do.

  It can be appreciated that when service configuration is performed with respect to service processing resources, multiple service processing resources can also be configured for the same service.

  Step 140: Receive a transfer rule used when a service transfer is to be made with respect to a service packet to be processed.

  Specifically, in one example, a network service processing device receives a transfer rule that is configured by a network server and is used when a service transfer is to be performed on a service packet to be processed. In addition, the network server configures transfer rules for the FPGA or NP.

  The network service processing apparatus may receive a transfer rule configured by another device other than the network server. In the present embodiment of the present invention, an example in which a network service processing apparatus receives a transfer rule configured by a network server will be used for description.

Furthermore, the transfer rule, identification information of the rule characteristic service processing service transfer with respect to each service packets that need to be performed is used when it should be done, regulatory characteristics and matches service processing resources And address information of service processing resources are stored. That is, only service packets that satisfy the rule characteristics can be transferred to the matching service processing resource using the address information of the service processing resource, and the service processing is performed by the service processing resource. Otherwise, discard processing is performed for service packets that do not satisfy the rule characteristics. By way of example and not limitation, a rule is specifically the IP address information, port number information, or VLAN ID identifier of a packet that needs to be serviced for that purpose, and the IP address information is also the source IP address information . It may be destination IP address information , and is not limited to that.

  In one example, the rule characteristics included in the forwarding rule include a plurality of IP address ranges of service packets for which service processing needs to be executed, identification information of service processing resources that match each IP address range, and service processing resources Address information. It can be seen that forwarding rules can be presented in tabular form. For example, a service packet having a first IP address range matches the identifier of the first service processing resource. In other words, when the IP address of the service packet to be processed received by the network service processing device falls within the first IP address range, the network service processing device is processed according to the address information of the service processing resource. The service packet to be transmitted is transmitted to the first service processing resource that matches the first IP address range.

  In this embodiment of the present invention, the time order relationship may not be limited between step 120, step 130, and step 140.

  Step 150: Receive a service packet to be processed sent by a user terminal, and the service processing resource is processed into a service processing resource using a transfer rule so as to perform service processing on the service packet to be processed Distribute service packets to be processed.

  Specifically, after executing Step 110 to Step 140 described above, the network service processing apparatus receives a service packet to be processed transmitted by the user terminal after performing a plurality of configurations.

  The network service processing device distributes the service packet to be processed using the forwarding rule and sends the service packet to be processed to the corresponding service processing resource, and the service processing resource receives the service packet to be processed Corresponding service processing is performed, thereby improving the efficiency of service packet processing.

Further, the step of distributing the service packet to be processed to the service processing resource using the transfer rule is specifically,
The network service processing device using a transfer rule to determine the amount of service processing resources that should perform service processing on the service packet to be processed.

  The step of determining the amount of service processing resources that should be serviced for the service packet to be processed is specifically a single service packet to be processed to perform the corresponding service processing. This is a step of determining that it is sufficient to have a service processing resource or a plurality of service processing resources are required to jointly perform corresponding service processing.

The step of distributing service packets to be processed to service processing resources is specifically:
When the network service processing device determines that the service processing should be performed on the service packet to be processed by one service processing resource, the network service processing device obtains the address information of the service processing resource from the transfer rule. Obtaining and directly distributing the service packet to be processed to the service processing resource determined according to the address information of the service processing resource, or the network service processing apparatus to the service packet to be processed multiple service processing resources When it is determined that the service processing should be performed, the network service processing apparatus determines the processing order for processing the service packet to be processed by the service processing resource, and transfers the processing according to the determined processing order. From the rules Obtaining address information of a number of service processing resources and distributing service packets to be processed to the first service processing resource according to the address information of the first service processing resource in the processing order.

  According to the network service processing method provided in the embodiment of the present invention, service processing is performed on service packets to be processed by a plurality of mutually isolated service processing resources that are divided FPGAs or NPs. Service processing is performed according to a configuration instruction with respect to a service processing resource, and service processing according to a transfer rule used when a service packet to be processed is to be serviced with respect to a service packet to be processed A forwarding rule is received so that service processing is performed on service packets that are distributed to the resources and the service processing resources are to be processed. Provides fast and flexible distribution of service packets and reduces the efficiency of distribution when the processor processes packets because the processor needs to identify and distribute each received packet when there is a relatively large number of packets The prior art problem of being done is avoided and the efficiency at which the server processes the packets is improved. In addition, the service processing resource can further have different service functions depending on the configuration, so that the function of the service processing resource can be changed at any time, and the service processing resource can be reconfigured. The efficiency with which service processing resources process service packets is improved.

Optionally, in this embodiment of the present invention, the service processing resource includes a step of collecting statistics on the service performance parameter at the same time when performing service processing on the service packet to be processed, Used to perform performance monitoring on service processing resources. By performing performance monitoring on service processing resources, the network operator to which the network server belongs can perform real-time monitoring on the performance of each service processing resource, and at the same time, each service processing to properly allocate network resources. Further adjustment of functions regarding resources can be performed. The specific steps are:
Collecting statistics on service performance parameters used by the FPGA or NP to perform performance monitoring on the service processing resource when the service processing resource performs service processing on the service packet to be processed;
FPGA or NP sends service performance parameters to a network server, where the service performance parameters send service performance parameters that are used as a basis for the network server to adjust the service load for service processing resources. And the steps to do.

Specifically, when each service processing resource performs corresponding service processing on a service packet to be processed, the FPGA or NP performs performance monitoring on each service processing resource according to the service configuration of each service processing resource. Collect statistics on the service performance parameters used to do it. By way of example and not limitation, the service performance parameters specifically include service traffic and service performance status for each service processing resource. Specifically, the service performance status refers to the remaining service packet processing capability after each service processing resource processes a service packet.

  In this embodiment of the invention, because the services configured for service processing resources are different, the FPGA or NP collects statistics on service performance parameters corresponding to the service processing resources. For example, statistics on the density of service packets received by each service processing resource, the size of service packets received by each service processing resource, the access density for accessing on-chip / off-chip memory, and the amount of data accessed Are collected according to services configured for service processing resources.

After the service performance parameter for each service processing resource is determined, the FPGA or NP sends the service performance parameter to the network server, which is the basis for the network server to adjust the service load for the service processing resource. Used as.

  Further, the network server determines whether the service processing resource needs to be changed according to the service performance parameter, and when the service processing resource needs to be changed, the network server Performs capacity expansion processing for resources or service reconfiguration processing for service processing resources. In this way, the network operator to which the network server belongs can perform real-time monitoring on the performance of each service processing resource, and at the same time, further function adjustment for each service processing resource can be performed in order to appropriately allocate network resources. It can be carried out.

Optionally, in this embodiment of the present invention, the network server can further perform a step of reconfiguring the service processing resource, so that the function of the service processing resource can be changed at any time, Processing resources can be reconfigured, which improves the efficiency with which each service processing resource processes service packets. The specific steps are:
Receiving a second configuration instruction when the service processing resource needs to have a different service function, the second configuration instruction including a second service configuration execution file, Receiving step;
Performing service reconfiguration on the service processing resource in accordance with the second service configuration execution file.

  Specifically, in one example, the network server can make functional adjustments for each service processing resource according to service performance parameters so as to achieve the objective of appropriately allocating network resources.

  When the service processing resource needs to have a different service function, the network service processing apparatus receives the second configuration instruction distributed by the network server including the second service configuration execution file. The network service processing device reconfigures the service processing resource having the service function according to the second service configuration execution file, thereby changing the function of the service processing resource, and each service processing resource processes the service packet. Efficiency is improved.

  The network service processing apparatus may further receive a second configuration command transmitted by another device other than the network server. In this embodiment of the present invention, an example in which the network service processing apparatus receives the second configuration command transmitted by the network server will be used for description.

  The second service configuration execution file included in the second configuration instruction and the first service configuration execution file included in the first configuration instruction perform different service functions so that the service processing resources have different functions. Can also be used.

  The second service configuration execution file included in the second configuration instruction includes a service configuration file and a service execution file. Specifically, the service configuration file is a programming file related to the service configuration, and the service execution file. Specifically, it is a programming file related to the performance of a service function by a service processing resource.

  The second configuration instruction further includes service processing resource identification information. The service processing device uses the identification information to determine a service processing resource on which service reconfiguration is to be performed. The identification information may be carried in the service configuration file, or the identification information is carried separately in the second configuration command.

  In step 110 of the present embodiment of the present invention, a plurality of mutually isolated service processing resources acquired by division are specifically empty service processing resources that do not have a service processing function or service processing functions. It is an empty service processing resource.

In one embodiment, when the service processing resource acquired by the division is an empty service processing resource, service configuration is performed for the empty service processing resource according to steps 120 to 130, and regarding the service processing resource according to these steps. Service reconfiguration may be performed.

In another embodiment, when the service processing resource acquired by the division is a non-empty service processing resource, service reconfiguration is performed for the non-empty service processing resource according to steps 120 to 130, and the service processing is performed according to these steps. Service reconfiguration may be performed again for the resource.

Embodiment 2.
Accordingly, an embodiment of the present invention further provides a network service processing apparatus. A mounting structure of the present apparatus is shown in FIG. 2, and the present apparatus is configured to realize the network service processing method of the first embodiment of the present invention described above. The apparatus includes a processor 10 and an FPGA 20 or NP20.

The processor 10 is configured to divide the FPGA or NP into a plurality of mutually isolated service processing resources 202 .

  The processor 10 receives the first configuration instruction including the first service configuration execution file, and performs the service processing on the service packet to be processed by the service processing resource. In accordance with the service processing resource.

  The FPGA 20 or NP20 is configured to receive transfer rules that are used when a service transfer is to be made with respect to the service packet to be processed.

  The FPGA 20 or NP20 receives the service packet to be processed transmitted by the user terminal, and uses the transfer rule to the service processing resource so that the service processing resource performs the service processing on the service packet to be processed. It is further configured to distribute service packets to be processed.

The processor 10 receives the second configuration instruction including the second service configuration execution file when the service processing resource needs to have different service functions,
It is further configured to perform service reconfiguration on the service processing resource according to the second service configuration execution file.

The processor 10 includes a service adaptation module 101 , and the FPGA 20 or NP20 includes a receiving side classification module 201 and service processing resources.

  Specifically, the service adaptation module 101 is configured to perform service configuration regarding the service processing resource according to the first service configuration execution file, and to perform service reconfiguration regarding the service processing resource according to the second service configuration execution file. .

  Specifically, the receiving side classification module 201 is configured to set a transfer rule on the receiving side classification module.

  Specifically, the service processing resource 202 is configured to set a transfer rule on the service processing resource.

Specifically, the receiving side classification module 201 determines the amount of service processing resources that should perform service processing on the service packet to be processed according to the transfer rule,
Distributing service packets to be processed to one determined service processing resource when it is determined that service processing should be performed on service packets to be processed, or multiple When the service processing resource is determined to perform service processing on the service packet to be processed, the processing order for processing the service packet to be processed by the service processing resource is determined and determined. The service packets to be processed are distributed to the first service processing resource in the processing order according to the processed processing order.

Service processing resource 202 receives the service packet to be processed sent by the receiver classification module,
Perform corresponding service processing on the service packet to be processed,
Determine the next level of service processing resources to be serviced for service packets to be processed according to the forwarding rules;
It is further configured to send the processed service packet to a next level service processing resource.

  The FPGA 20 or NP20 further includes a statistics collection module 203.

  The statistics collection module 203 collects statistics on service performance parameters used to perform performance monitoring on service processing resources when service processing is performed on service packets to be processed by the service processing resources. The parameter is configured to be sent to the service adaptation module.

  The service adaptation module 101 is further configured to send service performance parameters to the network server, and the service performance parameters are used as a basis for the network server to adjust the service load for service processing resources.

The FPGA 20 or NP20 is configured to receive a service packet processed by a service processing resource and output a service packet, and a transmission transmission module 204,
And further includes a transfer module 205 configured to transfer service packets between the receiving side classification module and the service processing resource, and further configured to transfer service packets between the plurality of service processing resources.

  The processor 10 further includes a driver module 102 configured to provide a unified interface for the service adaptation module.

  In this embodiment of the invention, the service adaptation module 101 can be located in a separate entity rather than in a processor.

The FPGA 20 or NP20 manages off-chip memory and provides an off-chip storage controller 206 configured to provide a unified access interface,
A service packet transmission / reception interface 207 configured to receive a service packet transmitted by an external device or to transmit a service packet to the external device;
An instruction transmission interface 208 configured to receive instructions sent by the network server or to receive instructions sent by a service adaptation module located in the driver module or processor.

Each service processing resource 202 included in the FPGA 20 or NP 20 has a one-to-one correspondence with each service adaptation module 101 included in the processor 10, that is, one service adaptation module 101 is configured and configured with respect to one service processing resource 202. It should be noted that processing is performed. Of course, in actual application, a plurality of service processing resources 202 may correspond to one service adaptation module 101. When the network service processing device includes at least one service processing resource 202, the network server distributes at least one first configuration instruction to at least one service adaptation module 101, each first configuration instruction being a first Includes service configuration executable files. Each service adaptation module 101 performs service configuration for the corresponding service processing resource 202. When performing service configuration, the service adaptation module 101 receives only one service processing resource 202 so that each service processing resource 202 processes only one service for the service packet to be processed when receiving the service packet. Configure one service. In addition, the configuration of the service processing resource 202 can be changed according to a different configuration command transmitted by the network server, for example, a second configuration command according to the above-described embodiment transmitted by the network server. The second configuration instruction includes a second service configuration executable so that the service function of the service processing resource 202 can be changed at any time and the service processing resource 202 can be reconfigured, thereby The processing efficiency of the service processing resource 202 is improved, and it is avoided that the function cannot be changed.

  According to the network service processing apparatus provided in the embodiment of the present invention, service processing is performed on service packets to be processed by a plurality of mutually isolated service processing resources that are divided FPGAs or NPs. Service processing is performed according to a configuration instruction with respect to a service processing resource, and service processing according to a transfer rule used when a service packet to be processed is to be serviced with respect to a service packet to be processed A forwarding rule is received so that service processing is performed on service packets that are distributed to the resources and the service processing resources are to be processed. Provides fast and flexible distribution of service packets and reduces the efficiency of distribution when the processor processes packets because the processor needs to identify and distribute each received packet when there is a relatively large number of packets The prior art problem of being done is avoided and the efficiency at which the server processes the packets is improved. In addition, the service processing resource can further have different service functions depending on the configuration, so that the function of the service processing resource can be changed at any time, and the service processing resource can be reconfigured. The efficiency with which service processing resources process service packets is improved.

Embodiment 3.
To assist in understanding this embodiment of the invention, the following is further described with reference to the accompanying drawings using specific embodiments. This embodiment does not constitute a limitation on the embodiment of the present invention.

  The following will be described in detail with reference to FIG. FIG. 3 is a signaling diagram of a process in which a single service processing resource performs network service processing according to an embodiment of the present invention.

  Specifically, as shown in FIG. 3, an example in which one service processing resource performs service processing on a service packet will be used. In accordance with the description of step 110 to step 140 of the embodiment of the present invention, corresponding configurations are performed with respect to the service processing resource and the receiving side classification module. After the configuration is complete, when the user terminal accesses the network, the receiving classification module receives the first service packet sent by the user terminal, and a unique attribute of the first service packet (e.g., Service packet IP address information, port number information, and VLANID identifier. In this case, the IP address information specifically includes, but is not limited to, source IP address information and destination IP address information). Identify whether it matches the service characteristics of the service (eg, NAT service) in the rule (eg, IP address information, port number information, and VLAN ID identifier of the packet for which the NAT service needs to be executed). If the unique attribute of the first service packet matches the service characteristic of the service in the transfer rule, the receiving classification module should perform service processing on the first service packet according to the transfer rule. (For example, NAT service processing resource) is determined. That is, distribution processing is performed on the first service packet. That is, it is determined that only a single service processing resource is required to perform the service processing corresponding to the first service packet.

  When it is determined that one service processing resource should perform service processing on the first service packet, the address information of the NAT service processing resource is obtained from the forwarding rule. The receiving side classification module performs an encapsulation process on the first service packet in order to acquire the second service packet (specifically, the encapsulation process includes the identification information of the NAT service processing resource and the NAT service). The processing service address information is encapsulated in the header of the first service packet), the second service packet includes NAT service processing resource identification information and NAT service processing resource address information. The receiving side classification module sends the second service packet to the forwarding module, and the forwarding module sends the second service packet to the NAT service processing resource according to the identification information and the address information. The NAT service processing resource receives the second service packet, performs the corresponding NAT service processing on the second service packet, and outputs the processed second service packet.

  At the same time when the NAT service processing resource performs service processing on the second service packet, the statistics collection module collects statistics on the service performance parameter of the NAT service processing resource and sends the service performance parameter to the service adaptation module. The service adaptation module sends service performance parameters to the network server, and the service performance parameters are used as a basis for the network server to adjust the service load for NAT service processing resources. When the NAT service processing resource needs to be changed, the network server performs capacity expansion processing on the NAT service processing resource or performs service reconfiguration processing on the NAT service processing resource according to the service performance parameter.

Embodiment 4.
To assist in understanding this embodiment of the invention, the following is further described with reference to the accompanying drawings using specific embodiments. This embodiment does not constitute a limitation on the embodiment of the present invention.

  The following will be described in detail with reference to FIGS. 4A and 4B. 4A and 4B are signaling diagrams of processes in which a plurality of service processing resources perform network service processing according to an embodiment of the present invention.

  Specifically, as shown in FIGS. 4A and 4B, an example in which two service processing resources perform service processing on a service packet will be used. In accordance with the description of step 110 to step 140 of the embodiment of the present invention, corresponding configurations are performed with respect to the service processing resource and the receiving side classification module. After the configuration is complete, when the user terminal accesses the network, the receiving classification module receives the first service packet sent by the user terminal, and a unique attribute of the first service packet (e.g., Service packet IP address information, port number information, and VLANID identifier. In this case, the IP address information specifically includes, but is not limited to, source IP address information and destination IP address information). Match the service characteristics of multiple services (eg PPPoE and DPI services) in the rule (eg IP address information, port number information, and VLANID identifiers of the packets for which PPPoE and DPI services need to be executed) Identify whether or not. If the unique attribute of the first service packet matches the service characteristics of those multiple services in the forwarding rule, the receiver classification module should perform service processing on the first service packet according to the forwarding rule Service processing resources (eg, PPPoE service processing resources and DPI service processing resources) are determined. That is, distribution processing is performed on the first service packet. That is, it is determined that a plurality of service processing resources are required to perform service processing corresponding to the first service packet.

  When it is determined that the two service processing resources should process the two services of the first service packet, the processing order of the PPPoE service processing resource and the DPI service processing resource is further determined, and the PPPoE service processing resource The address information and the address information of the DPI service processing resource are acquired from the transfer rule according to the processing order of the PPPoE service processing resource and the DPI service processing resource. The receiving side classification module performs an encapsulation process on the first service packet in order to obtain the second service packet (specifically, the encapsulation process is performed on the PPPoE to the header of the first service packet). Service processing resource identification information and DPI service processing resource identification information, service processing resource address information, and order information for service processing resources to process services)), the second service packet is , PPPoE service processing resource, service processing resource address information, DPI service processing resource identification information, and order information for the service processing resource to process the service. The receiving side classification module sends the second service packet to the transfer module, and the transfer module sends the second service packet to the first service processing resource indicated by the order information according to the identification information, the address information, and the order information. (E.g., PPPoE service processing resource). The PPPoE service processing resource receives the second service packet and performs a corresponding PPPoE service process on the second service packet. In addition, the PPPoE service processing resource determines that DPI service processing needs to be further performed on the second service packet according to the order information, and then the PPPoE service processing resource acquires the third service packet. In order to do so, the second service packet is encapsulated (specifically, the DPI service processing resource identification information, the DPI service processing resource address information, and the order information are sent to the second service packet). Encapsulating in the header of the packet). The third service packet includes identification information of the DPI service processing resource, address information of the DPI service processing resource, and order information. The PPPoE service processing resource sends the third service packet to the transfer module, and the transfer module sends the third service packet to the DPI service processing resource indicated by the identification information according to the identification information and the address information. The DPI service processing resource receives the third service packet, performs the corresponding DPI service processing on the third service packet, and outputs the processed third service packet.

  At the same time when the PPPoE service processing resource and the DPI service processing resource perform service processing on the second service packet and the third service packet, respectively, the statistics collection module relates to the service performance parameter of the PPPoE / DPI service processing resource. Statistics are collected separately and each service performance parameter is sent separately to the corresponding service adaptation module. Each service adaptation module sends service performance parameters to the network server separately, and the service performance parameters are used as a basis for the network server to adjust the service load for PPPoE / DPI service processing resources. When the PPPoE / DPI service processing resource needs to be changed, the network server performs capacity expansion processing for the PPPoE / DPI service processing resource or reconfigures the service for the PPPoE / DPI service processing resource according to the service performance parameter. Process.

The order information in the second service packet contains the order of only two service processing resources, and since one service of the second service packet has already been processed, the DPI service processing resource is processed third It can be understood that the service packet is output directly. In actual application, if a service packet needs to be processed by at least two service processing resources, the processing process is similar to that described above, and details are not repeated here.

Embodiment 5.
To assist in understanding this embodiment of the invention, the following is further described with reference to the accompanying drawings using specific embodiments. This embodiment does not constitute a limitation on the embodiment of the present invention.

  The following will be described in detail with reference to FIGS. 5A and 5B. 5A and 5B are another signaling diagrams of processes in which multiple service processing resources perform network service processing according to an embodiment of the present invention.

  In the fourth embodiment, when performing the encapsulation process on the first service packet, the receiving side classification module obtains the second service packet by identifying the identification information of the plurality of service processing resources and the plurality of services. Processing resource order information is encapsulated in the header of the first service packet, so the header of the second service packet is too large, which prevents packet transmission.

  In this embodiment of the invention, the header of the first service packet is simplified to facilitate packet transmission.

  Specifically, as shown in FIGS. 5A and 5B, an example in which two service processing resources perform service processing on a service packet will be used. In accordance with the description of step 110 to step 140 of the embodiment of the present invention, corresponding configurations are performed with respect to the service processing resource and the receiving side classification module. After the configuration is complete, when the user terminal accesses the network, the receiving classification module receives the first service packet sent by the user terminal, and a unique attribute of the first service packet (e.g., Service packet IP address information, port number information, and VLANID identifier. In this case, the IP address information specifically includes, but is not limited to, source IP address information and destination IP address information). Match the service characteristics of multiple services (eg PPPoE and DPI services) in the rule (eg IP address information, port number information, and VLANID identifiers of the packets for which PPPoE and DPI services need to be executed) Identify whether or not. If the unique attribute of the first service packet matches the service characteristics of those multiple services in the forwarding rule, the receiver classification module should perform service processing on the first service packet according to the forwarding rule Service processing resources (eg, PPPoE service processing resources and DPI service processing resources) are determined. That is, distribution processing is performed on the first service packet. That is, it is determined that a plurality of service processing resources are required to perform service processing corresponding to the first service packet.

When it is determined that the two service processing resources should process the two services of the first service packet, the processing order of the PPPoE service processing resource and the DPI service processing resource is further determined, and the PPPoE service processing resource The address information and the address information of the DPI service processing resource are acquired from the transfer rule according to the processing order of the PPPoE service processing resource and the DPI service processing resource. The receiving side classification module performs the encapsulation process on the first service packet to acquire the second service packet (specifically, the encapsulation process is performed according to the service processing order of the two service processing resources). The identification information and address information of the first service processing resource in the service processing order, for example, the identification information and address information of the PPPoE service processing resource are encapsulated in the header of one service packet). The second service packet includes PPPoE service processing resource identification information and address information. The receiving side classification module sends the second service packet to the transfer module, and the transfer module transfers the second service packet to the PPPoE service processing resource. The PPPoE service processing resource performs the corresponding PPPoE service processing for the second service packet, and the unique attribute of the second service packet matches the service characteristics of multiple services (for example, DPI service) in the service forwarding rule table Identify whether to do. If the unique attribute of the second service packet matches the service characteristics of those multiple services in the service forwarding rule table, the PPPoE service processing resource further needs to perform service processing on the second service packet. A service processing resource (for example, DPI service processing resource) is determined, and the DPI service processing resource acquires identification information and address information of the DPI service processing resource from the transfer rule. If the unique attribute of the second service packet does not match the service characteristics of those services in the service forwarding rule table, the second service packet is output. The PPPoE service processing resource performs an encapsulation process on the second service packet in order to acquire the third service packet (specifically, the encapsulation process includes identification information and address information of the DPI service processing resource). The third service packet includes identification information and address information of the DPI service processing resource. The PPPoE service processing resource sends the third service packet to the transfer module, and the transfer module sends the third service packet to the DPI service processing resource indicated by the identification information according to the identification information and the address information. The DPI service processing resource receives the third service packet, the DPI service processing resource performs the corresponding DPI service processing on the third service packet, and the unique attribute of the third service packet is in the service forwarding rule table. Identify if it matches the service characteristics of multiple services. If the unique attribute of the third service packet matches the service characteristics of those multiple services in the service forwarding rule table, the DPI service processing resource further needs to perform service processing on the third service packet. The service processing resource to be determined is determined. If the unique attribute of the third service packet does not match the service characteristics of the plurality of services in the service forwarding rule table, the third service packet is output.

  At the same time when the PPPoE service processing resource and the DPI service processing resource perform service processing on the second service packet and the third service packet, respectively, the statistics collection module relates to the service performance parameter of the PPPoE / DPI service processing resource. Statistics are collected separately and each service performance parameter is sent separately to the corresponding service adaptation module. Each service adaptation module sends service performance parameters to the network server, and the service performance parameters are used as a basis for the network server to adjust the service load for PPPoE / DPI service processing resources. When the PPPoE / DPI service processing resource needs to be changed, the network server performs capacity expansion processing for the PPPoE / DPI service processing resource or reconfigures the service for the PPPoE / DPI service processing resource according to the service performance parameter. Process.

  Since the transfer rules are configured separately for each service processing resource, each service processing resource determines the service processing resource that will process the service packet next, thereby reducing the number of bytes in the header of the service packet, It can be understood that the flexible transmission of the service is facilitated and the processing efficiency of each service processing resource is improved.

  In actual application, if a service packet needs to be processed by at least two service processing resources, the processing process is similar to that described above, and details are not repeated here.

  One skilled in the art can further appreciate that the units and algorithm steps may be implemented by electronic hardware, computer software, or a combination thereof, in combination with the examples described in the embodiments disclosed herein. . In order to clearly describe the compatibility between hardware and software, the above generally describes the configuration and steps of each example according to function. Whether each function performs hardware or software depends on the particular application and design constraints of the technical solution. Those skilled in the art can use various methods to implement the described functions for each application, but such embodiments should not be considered beyond the scope of the present invention.

  The method or algorithm steps described in the embodiments disclosed herein may be implemented by hardware, software modules executed by a processor, or a combination thereof. Software modules can be random access memory (RAM), memory, read only memory (ROM), electrically programmable ROM, electrically erasable ROM, registers, hard disk, removable disk, CD-ROM, or in the art It can be placed on any other known form of storage medium.

  In the foregoing specific embodiments, the objects, technical solutions, and beneficial effects of the present invention are described in detail. It should be understood that the foregoing descriptions are merely specific embodiments of the present invention and are not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, or improvement made without departing from the spirit and principle of the present invention shall fall within the protection scope of the present invention.

10 processor
20 FPGA or NP
101 Service adaptation module
102 Driver module
201 Receiver classification module
202 Service processing resources
203 Statistics collection module
204 Transmission module
205 Transfer module
206 Off-chip storage controller
207 Service packet transmission / reception interface
208 Command transmission interface

Claims (11)

A network service processing method,
Dividing the field programmable gate array FPGA or network processor NP into a plurality of mutually isolated service processing resources;
Receiving a first configuration instruction, wherein the first configuration instruction includes a first service configuration execution file, and receiving the first configuration instruction;
Performing a service configuration on the service processing resource according to the first service configuration execution file so that the service processing resource performs a service process on a service packet to be processed;
Receiving a transfer rule used when a service transfer is to be made with respect to the service packet to be processed;
Receiving the service packet to be processed transmitted by a user terminal and using the transfer rule to the service processing resource so that the service processing resource performs service processing on the service packet to be processed Distributing a service packet to be processed. A network service processing method. The step of distributing the service packet to be processed to the service processing resource using the transfer rule specifically includes:
Using the forwarding rule to determine the amount of service processing resources to be serviced for the service packet to be processed;
When it is determined that one service processing resource is to perform service processing on the service packet to be processed, the service packet to be processed is distributed to the determined one service processing resource. Step, or when a plurality of service processing resources are determined to perform service processing on the service packet to be processed, the service processing resource for processing the service packet to be processed The network service processing method according to claim 1, further comprising: determining a processing order and distributing the service packets to be processed to a first service processing resource in the processing order according to the determined processing order. . The method
Collect statistics on service performance parameters used by the FPGA or the NP to perform performance monitoring on the service processing resource when the service processing resource performs service processing on the service packet to be processed And steps to
The FPGA or the NP sends the service performance parameter to a network server, the service performance parameter being used as a basis for the network server to adjust a service load for the service processing resource; The network service processing method according to claim 1 or 2, further comprising: transmitting the service performance parameter. The method
Receiving a second configuration instruction when the service processing resource needs to have a different service function, wherein the second configuration instruction includes a second service configuration execution file; Receiving a command; and
The network service processing method according to claim 1, further comprising: performing a service reconfiguration with respect to the service processing resource in accordance with the second service configuration execution file. A network service processing device,
Including a processor and an FPGA or NP,
The processor is configured to divide the FPGA or the NP into a plurality of mutually isolated service processing resources;
The processor receives the first configuration instruction including the first service configuration execution file, and performs the service processing on the service packet to be processed by the service processing resource. Further configured to perform service configuration for the service processing resource according to an executable file;
The FPGA or the NP is configured to receive a transfer rule used when a service transfer is to be made with respect to the service packet to be processed;
The FPGA or the NP receives the service packet to be processed transmitted by a user terminal, and sets the transfer rule so that the service processing resource performs service processing on the service packet to be processed. Further configured to use to distribute the service packet to be processed to the service processing resource;
Network service processing device. The processor receives a second configuration instruction including a second service configuration execution file when the service processing resource needs to have different service functions;
6. The network service processing apparatus according to claim 5, further configured to perform service reconfiguration with respect to the service processing resource in accordance with the second service configuration execution file. The processor includes a service adaptation module, and the FPGA or the NP includes a receiving side classification module and the service processing resource,
Specifically, the service adaptation module performs service configuration with respect to the service processing resource according to the first service configuration execution file, and performs service reconfiguration with respect to the service processing resource according to the second service configuration execution file. Configured,
The receiver classification module is specifically configured to set the forwarding rule on the receiver classification module;
The network service processing apparatus according to claim 5 or 6, wherein the service processing resource is specifically configured to set a transfer rule on the service processing resource. Specifically, the receiving side classification module determines an amount of service processing resources that should perform service processing on the service packet to be processed according to the forwarding rule,
When one service processing resource is determined to perform service processing on the service packet to be processed, the service packet to be processed is distributed to the determined one service processing resource. Or a process for the service processing resource to process the service packet to be processed when it is determined that a plurality of service processing resources should perform service processing on the service packet to be processed 6. The network service processing according to claim 5, wherein the network service processing is configured to determine an order and distribute the service packet to be processed to a first service processing resource in the processing order according to the determined processing order. apparatus. The service processing resource is:
Receiving the service packet to be processed sent by the receiver classification module;
Performing corresponding service processing on the service packet to be processed;
Determining a next level of service processing resources to be serviced for the service packet to be processed according to the forwarding rules;
9. The network service processing device according to claim 8, further configured to send a service packet to be processed acquired after the service processing to the next level service processing resource. The FPGA or the NP further includes a statistics collection module;
The statistics collection module collects statistics on service performance parameters used to perform performance monitoring on the service processing resource when the service processing resource performs service processing on the service packet to be processed. Configured to send the service performance parameters to the service adaptation module;
The service adaptation module is further configured to send the service performance parameter to a network server, the service performance parameter used as a basis for the network server to adjust a service load for the service processing resource. The network service processing device according to claim 5, wherein the network service processing device is used. The FPGA or the NP is
A transmission module configured to receive a service packet processed by the service processing resource and to output the service packet;
A transfer module configured to transfer the service packet between the receiving side classification module and the service processing resource, and further configured to transfer the service packet between the plurality of service processing resources; 6. The network service processing device according to claim 5, further comprising:

Images